Run a job on the network — no code required

Work can be submitted to the Open Compute Network from a browser or from the Python SDK. Common capabilities include reading scanned PDFs page by page in any script, building a vector index over your own documents, classifying and extracting fields from unstructured text, transcoding media, and fanning simulations across many machines at once. Every submission carries a budget ceiling that the network enforces rather than treats as a hint.

Scanned documents are read one page per machine, with per-page confidence reported honestly.
A vector index built on the network is a file you keep; it is not held here.
Every call takes a budget in OCN, which is a ceiling and not an estimate.
Identical work that somebody has already published is served from cache without charge.

Elsewhere on this site

The Cube Floor — The Open Compute Network. The live figures on this page are loaded from https://api.thecubefloor.com, which is public and needs no credentials to read.